A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Classification Techniques in Machine Learning
Machine learning approaches can be primarily grouped into three categories: supervised learning, unsupervised learning and semi‐supervised learning. Among these, supervised learning is widely adopted, where a model is trained using labeled data, consisting of both input attributes and associated output labels. Like regression, classification belongs to supervised learning techniques. The objective is to build models that can predict the class label of unseen data, after being trained on a representative set of labeled data. This chapter discusses classification techniques, such as logistic regression, support vector machines, and ensemble methods, and their strengths and limitations, as well as important technical aspects like pre‐processing, evaluation metrics and cross‐validation. Finally, it presents the basic aspects of active learning (AL) in classification. AL is particularly useful for classification tasks when labeled data are scarce or expensive to obtain, which is often the case in materials science, either experimentally of numerically, and geophysics.
Classification Techniques in Machine Learning
Machine learning approaches can be primarily grouped into three categories: supervised learning, unsupervised learning and semi‐supervised learning. Among these, supervised learning is widely adopted, where a model is trained using labeled data, consisting of both input attributes and associated output labels. Like regression, classification belongs to supervised learning techniques. The objective is to build models that can predict the class label of unseen data, after being trained on a representative set of labeled data. This chapter discusses classification techniques, such as logistic regression, support vector machines, and ensemble methods, and their strengths and limitations, as well as important technical aspects like pre‐processing, evaluation metrics and cross‐validation. Finally, it presents the basic aspects of active learning (AL) in classification. AL is particularly useful for classification tasks when labeled data are scarce or expensive to obtain, which is often the case in materials science, either experimentally of numerically, and geophysics.
Classification Techniques in Machine Learning
Stefanou, Ioannis (author) / Darve, Félix (author) / JAKSE, Noel (author)
Machine Learning in Geomechanics 1 ; 117-144
2024-10-25
28 pages
Article/Chapter (Book)
Electronic Resource
English
Concrete Classification Using Machine Learning Techniques
| Springer Verlag | 2022
Concrete Classification Using Machine Learning Techniques
| TIBKAT | 2022
CPT-Based Soil Classification through Machine Learning Techniques
| British Library Conference Proceedings | 2023